Kimeme

Kimeme
Developer(s) Cyber Dyne s.r.l
Stable release
1.0 / november, 2011
Development status Active
Operating system Cross-platform
Type Technical computing
License Proprietary
Website

Kimeme is an open platform for multi-objective optimization and multidisciplinary design optimization. It is intended to be coupled with external numerical software such as Computer Aided Design (CAD), Finite Element Analysis (FEM), Structural analysis and Computational Fluid Dynamics tools. It has been developed by Cyber Dyne Srl and provides both a design environment for problem definition and analysis and a software network infrastructure to distribute the computational load.

History

Cyber Dyne has been founded in 2011 as a research startup to transfer the knowledge of its founders in the field of numerical optimization and computational intelligence methods into a commercial product.

Features

The problem definition workflow is based on the data flow paradigm. Several icons can be interconnected in order to describe the flow of data from the design variables to the desired objectives and constraints. Input/output nodes can be used to calculate any part of the objective computation, using internal or external numerical procedures. Any of these procedures can be distributed over the LAN, exploiting all the available computational resources. The optimization core is open, and using the memetic computing (MC) approach, which is an extension of the concept of memetic algorithm, the user can define its own optimization algorithm as a set of independent pieces of code called "operators". Operators can be implemented either in Java or Python.

Algorithm design

In mathematical folklore, the no free lunch theorem (sometimes pluralized) of David Wolpert and William G. Macready appears in the 1997 "No Free Lunch Theorems for Optimization."[1] This mathematical result states the need for a specific effort in the design of a new algorithm, tailored to the engineering problem to be optimized. Kimeme handles the design and experimentation of new optimization algorithms through the new paradigm of memetic computing, which is a subject of computational intelligence which studies algorithmic structures composed of multiple interacting and evolving modules (memes). [2]

Design of experiments

Different strategies are available, including random generator sequences, Factorial DOEs, Orthogonal and Iterative Techniques, as like as D-Optimal or Cross Validation. Monte Carlo and Latin hypercube are available for robustness analysis.

Sensitivity analysis

Local Sensitivity as correlation coefficients and partial derivatives can be used only if the correlation between input and output is linear. If the correlation is nonlinear, the global sensitivity analysis has to be used based on the variance-relationship between input and output distribution as Sobol index. With sensitivity analysis, the system complexity can be reduced and the cause-effect chain can be explained.[3][4]

Multi-objective optimization

In the development process of technical products, there are usually several evaluation goals or criteria to be met, e.g. low cost, high quality, low noise etc. These criteria often conflict each other, in the sense that the minimization of one entails the maximization of at least another one. Design parameters have to be found in order to find the best trade-off among criteria. Unlike the single-objective case, in multi-objective optimization there is not a unique solution, but rather a Pareto optimal solution frontier. Multi-objective optimization aims at finding all Pareto solutions automatically with a single run.

See also

External links

References

  1. ↑ Wolpert, D.H., Macready, W.G. (1997), "No Free Lunch Theorems for Optimization," IEEE Transactions on Evolutionary Computation 1, 67. http://ti.arc.nasa.gov/m/profile/dhw/papers/78.pdf
  2. ↑ Neri, F. & Cotta, C. 2011. "A primer on memetic algorithms". In "F. Neri, C. Cotta & P. Moscato (Eds.) Handbook of Memetic Algorithms", "Springer. Studies in Computational Intelligence".
  3. ↑ Saltelli, A., Chan, K. and Scott, E.M.: Sensitivity Analysis. John Willey & Sons Chichester, New York 2000
  4. ↑ Oakley J.E., O´Hagan A.: Probabilistic Sensitivity Analysis of Computer Models: a Bayesian Approach. Journal of the Royal Statistical Society, Series B, 66:751-769, 2004
This article is issued from Wikipedia - version of the 5/20/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.